Abstract The NIH BRAIN Initiative report emphasizes the importance of integrated technological approaches that combine multiple, complementary and groundbreaking capabilities in a single, highly advanced instrument. Optical methods have tremendous potential for integration, by combining massive recordings of neural activity with sophisticated, causal manipulations of cells' activity patterns, to both observe and test hypotheses about the mechanisms underlying large-scale brain activity. An instrument of this kind should ideally be applicable to multiple brain regions concurrently, to reveal how different areas intercommunicate, and capable of monitoring individual cells over the long-term. Further, to study the biological basis for mental states, the BRAIN Initiative seeks multi-scale recording technology that can, e.g. reveal the spiking dynamics of large numbers of individual cells, while simultaneously tracking population-level dynamics at larger spatial scales. To address these challenges, Stanford University and the Allen Institute for Brain Science will construct integrated optical instrumentation and transgenic reporter mice that are expressly designed to work together and that merge three capabilities, each of which is well beyond the present state-of-the-art when considered individually. We will construct a multi-beam, laser-scanning in vivo two-photon microscope that can track cellular spiking dynamics via Ca2+ imaging across a 16 mm2 area of tissue (i.e. sufficiently large to monitor multiple cortical areas simultaneously at cellular resolution in an awake mouse), image population-level voltage dynamics over the same area, and perform targeted manipulations of 500?1000 cells' activity levels by two- photon optogenetics. To reliably express optical indicator or actuator molecules across such large areas of brain tissue, we will also create transgenic reporter mice that express fluorescent optical voltage indicators or optogenetic actuator molecules targeted to the cell body membrane. To make these mice, we will use a novel genetic strategy that enables high expression levels in neuron-types that were hard to target in previous transgenic reporter lines. Our new method will yield double reporter mice allowing independent genetic control of two optical reporter molecules, such as a Ca2+ indicator and an opsin, so that these molecules can be targeted to two distinct cell types. Together, our reporter mice and optical instrumentation will provide unprecedented means to observe the brain's multi-scale network dynamics and test the causal linkages between large-scale neural activity patterns and animal behavior. If our work succeeds, it will be a `game-changer' for the study of large-scale brain circuitry, yielding crucial knowledge about how cells and the brain's global networks function normally and malfunction in disease. We expect that the data on multi-scale, brain area interactions provided by our instrumentation will lead to major conceptual advances in the understanding of how mammalian brains work.